Tobacco smoking is responsible for more than 90% of lung cancer cases, and yet the precise molecular alterations induced by smoking in lung that elaborate into cancer and impact survival have remained obscure. To identify the molecular signature associated with smoking, we conducted gene expression analysis using HG-U133A Affymetrix chips on 135 fresh-frozen tissue samples of adenocarcinoma and paired non-involved lung tissue from current, former, and never smokers, with biochemically validated smoking information. We used ANOVA analysis adjusted for potential confounders, gene set enrichment analysis, and GO-functional classification, with the p-value less than 0.001, the fold change more than 1.5, and false discovery rate used for gene selection. We validated in the original tissue samples by RT-PCR, and confirmed our results in independent tumor and non-tumor tissues from two studies. We identified a gene expression signature characteristic of smoking. This signature was associated with adenocarcinoma development and survival. Cell cycle gene changes, particularly those in the mitotic spindle formation (e.g., NEK2, TTK, BIRC5), strongly differentiated both smokers from non-smokers and early stage tumor from non-involved lung tissue, consistent with an important role for this pathway in smoking-related lung carcinogenesis. These changes persisted even decades after smoking cessation. NEK2 and TTK expression in the non-involved lung tissue was also associated with a 3-fold increased risk of mortality from lung cancer in smokers. Our work provides novel insights into how smoking induces lung cancer and affects survival, and identifies targets for chemoprevention and treatment against lung cancer in smokers.